This paper presents a stability theorem for a class of nonlinear fractional-order time-variant systems with fractional order α(1<α<1) by using the Gronwall-Bellman lemma. Based on this theorem, a sufficient condition for designing a state feedback controller to stabilize such fractional-order systems is also obtained. Finally, a numerical example demonstrates the validity of this approach.
1. Introduction
Recently, fraction-order system or system containing fractional derivatives and integrals has been studied widely [1–5]. It was found that many systems in interdisciplinary fields could be elegantly described with the help of fractional derivatives and integrals, such as viscoelastic system, dielectric polarization, electrode-electrolyte polarization, and electromagnetic waves [2–4]. In fact, real word processes generally or most likely are fractional-order systems. Moreover, fractional-order controllers [6, 7] have so far been implemented to enhance the robustness and the performance of the closed loop control system.
The problem of stability is a very essential and crucial issue for control systems including fractional-order system. Very recently, the stability problem of fractional-order system has been investigated both from an algebraic and from an analytic point of view [8–11]. By analyzing the characteristic equation of the Jacobian matrix, an asymptotically stable critical was proposed in [12]. As a way of efficiently solving the robust stability and stabilization problem, the linear-matrix-inequality (LMI) approach was presented [13–15] that provided the sufficient condition and the designing method of stabilizing controllers for fractional-order system. Note that the existing LMI-based control methods for fractional-order system only focus on the linear system but not on the case of nonlinear system. To account this problem, based on the generalization of Gronwall-Bellman lemma, the analytical stability conditions and state feedback stabilization problem of nonlinear affine fractional-order system have been investigated in [16–18]. In [19], by using of Mittag-Leffler function, Laplace transform, and the generalized Gronwall inequality, a new sufficient condition ensuring local asymptotic stability and stabilization of a class of fractional-order nonlinear systems with fractional-order α(1<α<2) was proposed. Based on Lyapunov’s second method, a novel stability criterion for a class of nonlinear fractional differential system was presented in [20].
Motivated by the above mentioned works, the main purpose of this this paper is to consider the stability problem of a class of nonlinear fractional-order time-variant systems. The main contribution of this paper is as follows. First, the time-variant uncertainty was discussed for fractional-order nonlinear system. Second, by using Gronwall-Bellman lemma, a stability condition for such fractional-order time-variant systems is presented. The rest of this paper is organized as follows. In Section 2, the problem formulation and some preliminaries are presented. The main results are derived in Section 3. The efficiency of the approach is shown through an illustrative example in Section 4. Finally, some conclusions are drawn in Section 5.
Throughout this paper, Rn denotes an n-dimensional Euclidean space and Rn×m is the set of all n×m real matrices. The matrix norms are defined as ∥A∥=∑i,j=1n|aij|(i,j=1,2,…,n). ai,j is the element of matrix A, and lim-t→∞A(t)=(lim-t→∞ai,j(t))n×n, lim_t→∞A(t)=(lim_t→∞ai,j(t))n×n,(i,j=1,2,…,n).
2. Fractional Derivative and Preliminaries
In this paper, the following Caputo definition [1] is adopted for fractional derivative of order α for function f(t):
(1)Dαf(t)=dαf(t)dtα=1Γ(α-m)∫0tf(m)(τ)(t-τ)α+1-mdτ,
where m is an integer satisfying m-1<a≤m and Γ(·) is the well-known Gamma function. The Laplace transform of the Caputo fractional derivative (1) of order α is
(2)∫0∞Dαf(t)e-stdt=sαL{f(t)}-∑k=0m-1s-α-k+1f(k)(t)|t=0,
where s∈C denotes the Laplace operator. Note that upon considering all the initial conditions to be zero, (2) can be reduced to
(3)L{Dαf(t)}=sαL{f(t)}.
The two-parameter Mittag-Leffler function Eα,β(z), which plays a very important role in the fractional calculus, is introduced as follows.
Definition 1 (see [1]).
The two-parameter Mittag-Leffler function is defined by
(4)Eα,β(z)=∑k=0∞zkΓ(αk+β),α>0,β>0.
The Laplace transform of the Mittag-Leffler function is
(5)∫∞e-sttαk+β-1Eα,β(k)(αtα)dt=k!sα-β(sα-α)k+1,Re(s)>|α|1/α,
where Eα,β(k)=(dk/dtk)Eα,β.
To prove the main results in the next section, the following lemmas are needed.
Lemma 2 (see [1]).
If α<2, β is an arbitrary real number, η is a constant satisfying απ/2<η<min{π,πα}, and M>0 is a real constant, then
(6)|Eα,β(z)|<M1+|z|,η≤|arg(z)|≤π,|z|>0.
For the n-dimension matrix, one has the following lemma.
Lemma 3 (see [17]).
If A∈Cn×n and α<2, β is an arbitrary real number, η is such that απ/2<η<min{π,πα} and C>0 is a real constant, then
(7)∥Eα,β(A)∥<M1+∥A∥,η≤|argλi(A)|≤π,kkkkkkkkkkkkkkkkkkkkkkkkkki=1,2,…,n,
where λi(A) denotes the ith eigenvalue of matrix A and ∥A∥ denotes the l2-norm.
Lemma 4 (Gronwall-Bellman lemma [21]).
If
(8)x(t)≤h(t)+∫t0tk(s)x(s)ds,t∈[t0,T),
where all the functions involved are continuous on [t0,T), T≤∞, and k(t)≥0, then x(t) satisfies
(9)x(t)≤h(t)+∫t0th(s)k(s)exp[∫stk(u)du]ds,kkkkkkkkkkkkkkkkkkkkkkkkkkkkkkt∈[t0,T).
3. Stability Analysis of Nonlinear Fractional-Order Time-Variant Systems
Consider the n-dimensional nonlinear fractional-order time-variant system described by the following form:
(10)Dαx(t)=A(t)x(t)+f(t,x(t))+Bu(t),
where 0≤α<1, A(t)∈Rn×n is a time-variant regular matrix, and f(t,x(t)) is a nonlinear function of state x(t), which is Lebesgue measurable with respect to t on (0,∞). Let f(t,0)≡0, then x(t)=0 is a zero solution of nonlinear fractional-order time-variant system (10) with u(t)=0.
Based on the aforementioned definition and lemmas, we first give out the stability theorem for nonlinear fractional-order time-variant system (10).
Theorem 5.
For the nonlinear fractional-order system (10) with u(t)=0, assume that
lim¯t→∞A(t)=A-,lim_t→∞A(t)=A_, and ∥A--A_∥<b;
f(t,0)=0, and lim∥x∥→0(∥f(t,x(t))∥/∥x(t)∥)=0;
|arg(spec(A-))|>aπ/2, αρ(A-)>1,
then the zero solution of system (10) is asymptotically stable.
Proof.
By the assumption (i), it yields
(11)lim¯t→∞aij(t)=a-ij,lim_t→∞aij(t)=a_ij,(i,j=1,2,…,n),
that is, for any given 0<ε<(b-∥A--A_∥)/2n2, there exists tij≥t0>0, when t≥tij, it has
(12)aij(t)<a-ij+ε,aij(t)>a_ij-ε.
Setting T1=max{tij∣i,j=1,2,…,n}, when t≥T1, then
(13)a_ij-ε<aij(t)<a-ij+ε.
As a result,
(14)a_ij-ε<a-ij<a-ij+ε.
Without loss of generality, assume that |arg(spec(A-))|>aπ/2, then from (13) and (14), it obtains
(15)|aij(t)-a-ij|<|a-ij-a_ij+2ε|<|a-ij-a_ij|+2ε,(t≥T1).
It follows that,
(16)∑i,j=1n|aij(t)-a-ij|<∑i,j=1n|a-ij-a_ij|+∑i,j=1n2ε,(t≥T1),
that is
(17)∥A(t)-A-∥<∥A--A_∥+2n2ε<b,(t≥T1).
On the other hand, by the assumption (ii) of Theorem 5, for any given ε0>0, there exists T2≥t0>0, when t≥T2, it has
(18)∥f(t,x(t))∥<ε0∥x(t)∥.
Rewriting system (10) with u(t)=0 as
(19)Dαx(t)=A-x(t)+(A(t)-A-)x(t)+f(t,x(t)),
taking Laplace transform on the system (19), it obtains
(20)X(s)=(Isα-A-)-1(sα-1x(0)kkkkkkkkkkkk+L{(A(t)-A-)x(t)+f(t,x(t))}).
Then, taking Laplace inverse transform for (20) by using the inverse Laplace transform formula of the Mittag-Leffler function, it yields
(21)x(t)=Eα,1(A-tα)x(0)+∫0t(t-τ)α-1Eα,α(A-(t-τ)α)×((A(τ)-A-)x(τ)+f(τ,x(τ)))dτ.
Taking T*=max{T1,T2}, when t≥T*, we have
(22)∥x(t)∥≤∥Eα,1(A-tα)∥∥x(0)∥+∫0t∥(t-τ)∥α-1∥Eα,α(A-(t-τ)α)∥×(∥(A(τ)-A-)∥∥x(τ)∥+∥f(τ,x(τ))∥)dτ.
Substituting (17) and (18) into (22), it results
(23)∥x(t)∥≤∥Eα,1(A-tα)∥∥x(0)∥+∫0tC1∥(t-τ)∥α-1∥Eα,α(A-(t-τ)α)∥∥x(τ)∥dτ,
where C1=b+ε0.
Since the eigenvalues of A- satisfy the assumption (iii) of Theorem 5, then from Lemma 3 there exists a real constant C2>0 such that
(24)∥Eα,1(A-tα)∥≤C21+∥A-tα∥,∥Eα,1(A-(t-τ)α)∥≤C21+∥A-(t-τ)α∥.
Substituting (24) into (23), it gives
(25)∥x(t)∥≤C21+∥A-tα∥∥x(0)∥+C1C2∫0t∥(t-τ)∥α-11+∥A-(t-τ)α∥∥x(τ)∥dτ=C21+∥A-∥tα∥x(0)∥+C1C2∫0t∥(t-τ)∥α-11+∥A-∥(t-τ)α∥x(τ)∥dτ.
Setting
(26)h(t)=C21+∥A-tα∥∥x(0)∥,k(t)=(t-τ)α-11+∥A-∥(t-τ)α,
and using Gronwall-Bellman lemma, it follows from (25) that
(27)∥x(t)∥≤C21+∥A-∥tα∥x(0)∥+C1C2×∫0t∥x(0)∥1+∥A-∥τα·(t-τ)α-1(1+∥A-∥(t-τ)α)1-1/(α∥A-∥)dτ.
The integral in (27) equals the sum of the two parts
(28)∫0t/2∥x(0)∥1+∥A-∥τα·(t-τ)α-1(1+∥A-∥(t-τ)α)1-1/(α∥A-∥)dτ+∫t/2t∥x(0)∥1+∥A-∥τα·(t-τ)α-1(1+∥A-∥(t-τ)α)1-1/(α∥A-∥)dτ.
Since α<1 and (t-τ)≥τ when τ∈[0,t/2], we obtain
(29)∫0t/2∥x(0)∥1+∥A-∥τα·(t-τ)α-1(1+∥A-∥(t-τ)α)1-1/(α∥A-∥)dτ≤∫t/2t∥x(0)∥1+∥A-∥τα·τα-1(1+∥A-∥τα)1-1/(α∥A-∥)dτ.
Similarly,
(30)∫t/2t∥x(0)∥1+∥A-∥τα·(t-τ)α-1(1+∥A-∥(t-τ)α)1-1/(α∥A-∥)dτ≤∫t/2t∥x(0)∥1+∥A-∥(t-τ)α-1·(t-τ)α-1(1+∥A-∥(t-τ)α)1-1/(α∥A-∥)dτt-τ=s__∫0t/2∥x(0)∥sα-1(1+∥A-∥sα)(1+∥A-∥sα)1-1/(α∥A-∥)ds.
From (28) to (30), and with aρ(A-)>1, (27) gives
(31)∥x(t)∥≤C2∥x(0)∥1+∥A-∥tα+2C1C2∫0t/2∥x(0)∥sα-1(1+∥A-∥sα)2-1/(α∥A-∥)ds=2C1C2∥x(0)∥α∥A-∥-1+C2∥x(0)∥1+∥A-∥tα+2C1C2∥x(0)∥(1-α∥A-∥)(1+∥A-∥(tα/2))1-1/(α∥A-∥).
Now for arbitrarily small δ>0, it can be proved that
(32)∥x(0)∥<δ⟹∥x(t)∥<2C1C2δα∥A-∥-1+C2δ1+∥A-∥tαkkkkkkkk+2C1C2δ(1-α∥A-∥)(1+∥A-∥(tα/2))1-1/(α∥A-∥),
which implies stability of the zero solution. This completes the proof.
In particular, if limt→∞A(t)=A, the following corollary is easily derived.
Corollary 6.
For the nonlinear fractional-order system (10) with u(t)=0, assume that
limt→∞A(t)=A and |arg(spec(A))|>aπ/2, αρ(a)>1;
f(t,0)=0, and lim∥x∥→0(∥f(t,x(t))∥/∥x(t)∥)=0,
then the zero solution of system (10) is asymptotically stable.
Based on Corollary 6 and according to the stability theory of linear fractional-order system and the pole placement technique of the linear control theory (see, e.g., [16, 17, 22]), it is easy to get the stabilization theorem of nonlinear fractional-order system (10).
Corollary 7.
Assume that
limt→∞A(t)=A;
f(t,0)=0, and lim∥x∥→0(∥f(t,x(t))∥/∥x(t)∥)=0,
and if K is selected to, such that, |arg(spec(A+BK))|>aπ/2, then the nonlinear fractional-order system (10) can be stabilized by the state feedback controller u(t)=Kx(t).
4. Numerical Example
Consider the following fractional-order time-variant Lorenz system:
(33)Dαx1=a(t)(x2-x1)+u(t)Dαx2=bx1-x1x3-c(t)x2+u(t)Dαx3=x1x2-dx3,
where 0≤α<1. Lorenz system (33) can be rewritten as
(34)Dαx(t)=A(t)x(t)+f(t,x(t))+B(t)u(t),
where x(t)=[x1(t),x2(t),x3(t)]T and
(35)A(t)=(-a(t)a(t)0b-c(t)000-d),B(t)=(111),f(t,x(t))=(0-x1(t)x3(t)x1(t)x2(t)).
Setting a(t)=9+sint, b=28, c(t)=-8, d=8/3, α=0.90. To simulate the fractional order systems, we use Oustaloup’s recursive poles/zeros filter [3] (an integer order system) of order three to approximate the fractional operator 1/s0.9, which has an error about 2 dB in the frequency range ω=10-2 rad/s to ω=102 rad/s. We use this recursive poles/zeros filter of order three to approximate the fractional operator. The initial value of the system [(x1(0),x2(0),x3(0)]=(0.22,0.24,0.12). The commensurate fractional-order Lorenz system (34) is chaotic without the controller, which is shown in Figure 1.
Time response of the selected systems with u(t)=0.
Note that f(t,0)=0 and
(36)lim∥x∥→0∥f(t,x(t))∥∥x(t)∥=lim∥x∥→0(x1(t)x2(t))2+(-x1(t)x3(t))2x12(t)+x22(t)+x32(t)≤lim∥x∥→0(x1(t)x2(t))2+(-x1(t)x3(t))2x12(t)=lim∥x∥→0x22(t)+x32(t)=0.
To stabilize the fractional-order system (33), we select K=[1-5-1], which makes ∥A--A_∥=4 and |arg(A-+BK)|>απ/2, by the eigenvalues of (A-+BK) being (-20.3516,-8.1380,-0.1771). So the conditions of Corollary 7 are satisfied. Thus, the fractional-order system (33) can be stabilized by the state feedback controller u(t)=Kx(t). The simulation result (Figure 2) shows that it is asymptotically stable and its states converge to zero, which shows that the obtained theoretic results are feasible and efficient for the nonlinear fractional-order system.
Time response of the selected systems with the control input u(t)=Kx(t).
5. Conclusion
In this paper, based on the Gronwall-Bellman lemma and the property of fractional calculus, a stabilization theorem of a class of nonlinear fractional-order time-variant systems with fractional order α(0≤α<1) has been proven theoretically. Furthermore, the sufficient condition for designing a state feedback controller to stabilize such fractional-order systems was also obtained. Finally, a numerical example demonstrates the validity of this approach.
Conflict of Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
Acknowledgments
This research is supported by National Natural Science Foundation of China (61104072) and Research Fund of Hunan Provincial Education Department (14JJ2073).
PodlubnyI.1999New York, NY, USAAcademic PressMR1658022HilferR.2000SingaporeWorld Scientific Publishing10.1142/9789812817747MR1890104OrtigueiraM. D.MachadoJ. A. T.Fractional signal processing and applications20038311228522862-s2.0-014189268210.1016/S0165-1684(03)00181-6WangJ.ZhouY.WeiW.Optimal feedback control for semilinear fractional evolution equations in Banach spaces201261447247610.1016/j.sysconle.2011.12.009MR2910321ZhouY.JiaoF.Existence of mild solutions for fractional neutral evolution equations20105931063107710.1016/j.camwa.2009.06.026MR2579471RaynaudH.-F.ZergaïnohA.State-space representation for fractional order controllers20003671017102110.1016/S0005-1098(00)00011-XMR1829699LadaciS.LoiseauJ. J.CharefA.Fractional order adaptive high-gain controllers for a class of linear systems200813470771410.1016/j.cnsns.2006.06.009MR2381496HwangC.ChengY.-C.A numerical algorithm for stability testing of fractional delay systems200642582583110.1016/j.automatica.2006.01.008MR2207823AhnH.-S.ChenY.Necessary and sufficient stability condition of fractional-order interval linear systems200844112985298810.1016/j.automatica.2008.07.003MR2527230LiY.ChenY. Q.PodlubnyI.Stability of fractional-order nonlinear dynamic systems: lyapunov direct method and generalized Mitta-Leffler stability20092414291468YakarC.Fractional differential equations in terms of comparison results and Lyapunov stability with initial time difference201020101676285710.1155/2010/762857SunG.WangM.HuangL.ShenL.Generating multi-scroll chaotic attractors via switched fractional systems20113061183119510.1007/s00034-011-9272-2MR2846572TavazoeiM. S.HaeriM.A note on the stability of fractional order systems20097951566157610.1016/j.matcom.2008.07.003MR2488105SabatierJ.MozeM.FargesC.LMI stability conditions for fractional order systems20105951594160910.1016/j.camwa.2009.08.003MR2595931LuJ.-G.ChenG.Robust stability and stabilization of fractional-order interval systems: an LMI approach20095461294129910.1109/TAC.2009.2013056MR2532618N'DoyeI.ZasadzinskiM.RadhyN. E.BouazizA.Stabilization of a class of nonlinear affine fractional-order systems using generalizations of Bellman-Gronwall lemmaProceedings of the 17th Mediterranean Conference on Control and AutomationJune 2009Thessaloniki, GreeceMakedonia Palace324330WenX. J.WuZ. M.LuJ. G.Stability analysis of a class of non-linear fractional-order systems2008551111781183LazarevićM. P.SpasićA. M.Finite-time stability analysis of fractional order time-delay systems: gronwall's approach2009493-447548110.1016/j.mcm.2008.09.011MR2483650ChenL.HeY.ChaiY.WuR.New results on stability and stabilization of a class of nonlinear fractional-order systems201475463364110.1007/s11071-013-1091-5MR3164616ZhouX.-F.HuL.-G.LiuS.JiangW.Stability criterion for a class of nonlinear fractional differential systems201428252910.1016/j.aml.2013.09.007MR3128642JonesG. S.Fundamental inequalities for discrete and discontinuous functional equations19641214357MR0162069'doyeI. N.ZasadzinskiM.DarouachM.RadhyN.Robust controller design for linear fractional-order systems with nonlinear time-varying model uncertaintiesProceedings of the 17th Mediterranean Conference on Control and AutomationJune 2009Thessaloniki, GreeceMakedonia Palace821826